Data

Tools

Indexes

Applications

Experiments

Open Source Science

234-on-device-learning-for-model-personalization-with-large-scale-cloud-coordinated-domain-adaption

https://github.com/szu-advtech-2024/234-on-device-learning-for-model-personalization-with-large-scale-cloud-coordinated-domain-adaption

Science Score: 13.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ○
    .zenodo.json file
  • ○
    DOI references
  • ○
    Academic publication links
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (4.9%) to scientific vocabulary
Last synced: 10 months ago · JSON representation

Repository

Basic Info
  • Host: GitHub
  • Owner: SZU-AdvTech-2024
  • Default Branch: main
  • Size: 1000 Bytes
Statistics
  • Stars: 0
  • Watchers: 0
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Created over 1 year ago · Last pushed over 1 year ago
Metadata Files
Citation

https://github.com/SZU-AdvTech-2024/234-On-Device-Learning-for-Model-Personalization-with-Large-Scale-Cloud-Coordinated-Domain-Adaption/blob/main/ 

# Implementation MPDA with pytorch

This is my implementation of the paper MPDA(On-Device Learning for Model Personalization with Large-Scale Cloud-Coordinated Domain Adaption) with pytorch. 

## Download Dataset

Download [Movielens-20m](https://grouplens.org/datasets/movielens/20m/) to /data/MovieLens

Download [Amazon Electronics](https://jmcauley.ucsd.edu/data/amazon/) to /data/Amazon

## Config
First you need to config the root path in /config.yml

## MovieLens Dataset Preprocess
generate users with train json file and users with train and test data json file
```shell
nohup python -u scripts/preprocess/movielens/generate_user_with_train_and_test.py > ./log/generate_user_with_train_and_test.log 2>&1 &
```

generate user and item mapping
```shell
python scripts/preprocess/movielens/generate_mapping_file.py
```

generate recall item pairs
```shell
python scripts/preprocess/movielens/generate_recall_item_pairs.py
```

## Amazon Dataset Preprocess
generate rating.csv with raw_data.json
```shell
python scripts/preprocess/amazon/generate_raw_data.py
```


## Initial Model
train global model NCF on MovieLens
```shell
nohup python -u scripts/train_global_model.py -model=NCF -epochs=10 -dataset=movielens -device=cuda:2 > ./log/train_global_model.log 2>&1 &
```

train mask model on MovieLens
```
nohup python -u scripts/train_mask_model.py -device=cuda:2 > ./log/train_mask_model.log 2>&1 &
```

transfer model NCF
```shell
bash ./commands/ncf_movielens_50_random.sh
```

Owner

  • Name: SZU-AdvTech-2024
  • Login: SZU-AdvTech-2024
  • Kind: organization

GitHub Events

Total
  • Watch event: 1
  • Push event: 3
  • Fork event: 1
  • Create event: 3
Last Year
  • Watch event: 1
  • Push event: 3
  • Fork event: 1
  • Create event: 3